I think that this is great for the most part. I think that in process caching in C# can be tricky, especially dealing with concurrency because even through the access to the backing dictionary for creating and retrieving is safe (assuming the function delegate for creating the object doesn't do something interesting), you are returned a reference to an object in memory.
At this point if you have multiple threads doing stuff with that object you can obviously run into issues.
Anyway concurrency is hard. I keep thinking of writing a cache that uses something like https://github.com/ReubenBond/DeepCopy to return a copy of the object.
At this point if you have multiple threads doing stuff with that object you can obviously run into issues.
Anyway concurrency is hard. I keep thinking of writing a cache that uses something like https://github.com/ReubenBond/DeepCopy to return a copy of the object.